Up to now there is no standardized strategy in how to teach evolutionary algorithms. Surely the answer to this question depends on the teaching context and the target group. But in the field of scientific teaching it is essential to abstract from paradigm orientation to a more general concept of evolutionary computing and to the underlying foundation. At the University of Stuttgart we pursue th...

This paper compares the use of RGB and HSV histograms during the execution of an Evolutionary Algorithm. This algorithm generates abstract images that try to match the histograms of a target image. Three different fitness functions have been used to compare: the differences between the individual with the RGB histogram of the test image, the HSV histogram, and an average of the two histograms a...

Co-evolution (i.e. the evolution of two or more competing populations with coupled fitness) has several interesting features that may potentially enhance the power of adaptation of artificial evolution. In particular, as discussed by Dawkins and Krebs [2], competing populations may reciprocally drive one another to increasing levels of complexity by producing an evolutionary “arms race”. In thi...

We present a new method for proving lower bounds in evolutionary computation based on fitness-level arguments and an additional condition on transition probabilities between fitness levels. The method yields exact or near-exact lower bounds for LO, OneMax, and all functions with a unique optimum. All lower bounds hold for every evolutionary algorithm that only uses standard mutation as variatio...

0

This paper presents a multiobjetive approach to solve the Linear Shelf Space Allocation Problem (LiSSAP), which consists on allocating lengths of shelves in a given shop to specific products or groups of products. Previously we gave the first steps towards the development of a commercially viable tool that used evolutionary computation to address the problem; in this paper we introduce MELiSSA,...

An evolutionary programming algorithm with adaptivemutation operators based on L evy prob ability distribution is studied L evy stable distri bution has an in nite second moment Because of this L evy mutation is more likely to generate an o spring that is farther away from its parent than Gaussian mutation which is often used in evolu tionary algorithms Such likelihood depends on a parameter in...

Parameter control is still one of the main challenges in evolutionary computation. This paper is concerned with controlling selection operators on-the-fly. We perform an experimental comparison of such methods on three groups of test functions and conclude that varying selection pressure during a GA run often yields performance benefits, and therefore is a recommended option for designers and u...